self-evident because the potential gains appear to be substantial when compared to the presumed economic costs of various types of policy actions that are usually suggested. However, as illustrated in the preceding section, the discussion of the gain from declining fertility is usually carried out on the macro-economic level and in terms of average indices, such as income per capita. Hence the demonstration that "society" benefits may be less than automatically persuasive as far as individuals or individual families are concerned; collective action aimed at reducing fertility must necessarily operate through individuals or single families rather than through some mythical average man. It is unfortunate that so much of the discussion of the economics of fertility control tends to blur or assume away this elementary point.
If the policies proposed are to be both successful and economically sound, it would be necessary to be quite specific about the distribution of the benefits and the costs that are involved in particular courses of action. This section of the paper is devoted mainly to an outline of the considerations which seem to be pertinent to this issue, and which are neglected in the literature.
To clarify the possible loci of incidence of the benefits and costs in a fertility control program, it may be helpful as a first approximation to picture society as composed of self-perpetuating autonomous units, called families. It will then be useful to distinguish clearly, if only conceptually, between two types of economic justification for a given policy, applicable according to whether the economic consequences-costs and benefits—of any action to reduce fertility taken by an individual family are borne or enjoyed entirely by that family or whether such actions involve certain externalities, i.e., impose burdens or confer benefits on other families as well.
Fertility Control in the Absence of Externalities
If all consequences of fertility decisions remained within the boundaries of the family, the economics of population policy under the usual, if somewhat vague, assumption of a democratic society appear rather simple. Families should be left to judge what they consider best for themselves, and society should accept the decisions of individual families with strict neutrality. The argument that "society" would be better off if only parents had fewer children is as meaningless as to say that society would be better off if only more people worked on Sundays. Clearly the presumed gains that would have accompanied lower fertility would have been by definition derived from the gains concentrated in the very families responsible for that lower fertility. If, in fact, families opted for a higher number of children, ipso facto that option is revealed as preferred. Nevertheless, a policy of pure laissez-faire does not necessarily follow. Our autonomous families do not live in isolation; they